我尝试在Apache Flink中处理后从1 KDS写入到另一个KDS。我使用Zeppelin notebook创建了以下查询的sink表：

%flink.ssql

CREATE TABLE seller_revenue (
    seller_id    VARCHAR,
    window_end   TIMESTAMP,
    sales        DOUBLE
) 
WITH (
    'connector'           = 'kinesis',
    'stream'              = 'seller_stream_window',
    'aws.region'          = 'ap-south-1',
    'scan.stream.initpos' = 'LATEST',
    'format'              = 'json'
)

字符串
然后我写数据使用以下

%flink.ssql(parallelism=1)

INSERT INTO seller_revenue
SELECT
    seller_id,
    TUMBLE_END(proctime, INTERVAL '30' SECONDS) AS window_end,
    SUM(product_quantity * product_price) AS sales
FROM seller_sales
GROUP BY
    TUMBLE(proctime, INTERVAL '30' SECONDS),
    seller_id

型
但出现以下错误：

Unable to create a sink for writing table 'hive.flink2.seller_revenue'.
Caused by: org.apache.flink.table.api.ValidationException: Unsupported options found for 'kinesis'.

Unsupported options:

scan.stream.initpos

型
有人能帮忙解决吗？
我尝试删除不支持的选项scan.stream.initpos，但没有数据被写入后，这一点。